From: "Peter Trei" <trei@process.com>
Date: Wed, 27 Nov 1996 10:56:38 -0800 (PST)
To: cypherpunks@toad.com
Subject: NIST seeks DES replacement.
Message-ID: <199611271856.KAA12134@toad.com>
MIME-Version: 1.0
Content-Type: text/plain


See:

http://www.fcw.com/pubs/fcw/1111/des.htm

Federal Computer Week, Nov 11, 1996.

"DES set for overhaul." by HEATHER HARRELD

The content is kind of thin, but it amounts to a statement that
NIST is preparing a Federal Register request for a successor to
DES. Most the article concerns the termendous hassle replacing
DES will be for government agencies (FCW is a trade rag for people
working with computers in the Federal and State sectors - I used to
get it when I was at MITRE.)

There's a truely clueless comment at the end, where Mike Schwartz
of "Prime Factors Inc., an Oregon-based security firm" is quoted
as saying "...DES shows no signs of weakening."

-----------------------------

While 3DES-EDE is the obvious replacement, it would be far from an
easy switch, since there are a huge number of fielded devices for 
handling the 64 bit single DES keys.

I wonder how good a drop-in replacement could be made
if the goal was to NOT have to replace the key handling 
infrastructure - just replace a single software module or chip.

If we use the bits currently devoted to parity in DES keys for
actual key data, we gain 8 bits, or a factor of x256 in the keyspace.

If we further complicate the key schedule setup, so it's much slower 
than DES in both hardware and software (lots of rotates, multiplies, 
state, etc), we can make life a lot more difficult for brute force attacks.

Maybe some version of Blowfish, using the same key twice - I have
not studied Blowfish, so I don't know if this introduces an obvious
weakness.

Happy Thanksgiving!

Peter Trei
trei@process.com

Peter Trei
Senior Software Engineer
Purveyor Development Team                                
Process Software Corporation
http://www.process.com
trei@process.com